Gold Shares: The Mystery of the Flatlining XAU

Until recently, the greatest threat to investors in gold shares was commonly perceived to be toxic hedge books. The May 1, 2007 proclamation by Barrick that it had reduced its hedge book in the first quarter by two million ounces, and at a substantial loss, is a fitting eulogy to the folly of hedging. It seems inconceivable that any management today would defend the practice in light of high profile hedging disasters, the under performance of mining shares with large hedge books, and the decade long drumbeat of investor displeasure.

The overriding investment risk for investors in gold mining shares today is neither geopolitical, nor geological nor gold price related. It is the danger of value destruction stemming from unwise financing decisions including equity placements and merger activity. Laxity in financial discipline is widespread in the industry, far more so than the speculative practice formerly known as hedging. For this reason, the historical and potential damage from shareholder dilution is equal to or greater than hedging the gold price.

The CEO of a major gold company recently voiced a mild complaint about the success of the gold ETF (GLD) to me. The gist of his comment was that GLD was siphoning away potential buyers for shares of gold mining stocks. He felt that In the absence of the ETF, shares of his company would probably be higher. I suggested that the real culprit might be excessive issuance of paper by the gold mining industry. I asked him if he could provide production and reserve growth in terms of ounces per share for his company right then and there. He was unable to give me an immediate answer and said that he would get back to me.

Barry Cooper, CIBC's eminent gold analyst, has noted that over the past ten years, the XAU (Philadelphia Stock Exchange Gold and Silver Sector Index), has made essentially no progress despite the 50% rise in the gold price and the doubling in the silver price over the same period. Some might say the culprit is higher costs which have outdistanced the gold price, especially energy. This argument has merit but it does not explain the whole story. It is true that gold has lagged other commodity prices over the last ten years (see chart).

Share issuance is the more compelling and revealing explanation. Aggregate shares outstanding have grown from 1.6 billion to 5 billion in the ten years. This means that the annual ounces of production per share have declined from 0.00814 to 0.0067 or nearly 18%. Reserves per share have shown a modest increase of 6% over the ten years.

The XAU is the most widely used performance benchmark for the gold sector. It has been around since 1983, and like all stock indices, it has not escaped the irresistible urge of those who oversee them to tinker with or refine their content. The XAU's gross market cap has increased from $29.2 billion at 12/31/96 to $130 billion. The number of underlying companies has increased from eleven to sixteen. Ounces of gold equivalent (silver ounces converted to gold at 52.3x) production have increased from 13.2 mm to 33.6 mm, or 154%. Gold equivalent ounces of reserves have grown from 189.4 million to 619.6 million or 227%.

Notwithstanding this apparent progress, shares outstanding for the underlying companies in the index have more than tripled. In this lies the principal explanation for the inability of the XAU share index to make any progress despite the rising gold price. All of the backup information for the ten year comparison, and then some, is contained in an appendix, prepared by Ting Haw Tu, our capable research analyst.

One exception to this line of criticism, of course, would be stock splits. In that case, share counts increase but shareholders own the same percentage of the company post split. We do not believe stock splits have been a major factor in the swelling count of shares outstanding. Randgold did declare a two for one split, but its overall share count is a paltry 68 million. AngloGold has declared a two for one split, but again, the share count of 275 million does not explain the 200% rise in shares over the period. Goldcorp had a two for one split in 1996 and a second one in 2002, after which shares outstanding were 182 mm. On the other hand, Kinross has had two reverse splits, 1 for 3 and 1 for 100. Freeport McMoran has had several share buybacks, a positive event for shareholders.

It goes without saying that mining for precious metals is a very capital intensive business. Returns on capital have shown considerable improvement since gold bottomed in 1999, but they could not be considered robust (see chart).

Few mines, and fewer companies, sport returns on capital comparable to Microsoft or other paragons of profitability. Commodity cost inflation has driven capital and operating costs upward at a quicker pace than the gold price. For all of these reasons, the need to resort to external equity financing is understandable and justifiable.

Nevertheless, the gold mining industry enjoys a distinct advantage from other extractive industries when it comes to external financing. Its cost of capital is considerably lower. For example, the current composite p/e ratio for trailing twelve months earnings for the 16 companies comprising the XAU is 28.3x. In contrast, leading base metal companies trade at 11.4x and major oil companies for 10.4x.

The explanation lies in the willingness of investors to value gold shares based on their "optionality" to variations in the gold price. Senior gold mining shares are in effect very long dated options on the gold price. When investors are bullish on the gold price, they clamor for such paper. The industry and its bankers rarely disappoint them. It is ironic that the scarcity of gold, which makes it precious, and which forms the very basis for the high value placed on gold shares by investors, can be so easily cancelled out or even debased by excessive and sometimes capricious issuance.

It would be preferable if management and its financial advisors regarded yet to be issued shares as precious as the metal yet to be mined, and accordingly consented to increases in the share count only with the greatest reluctance. With this in mind, we offer a few modest and (hopefully) constructive suggestions:

1. The metrics of ounces per share, both production and reserves, should be included in all communications to shareholders, and especially investor presentations. Information should be provided over a five year period.
2. Investor presentations should also contain historical and projected information for return on capital. This includes project specific information as well as corporate level data.
3. A few slides containing drill hole and geological data will not be missed and can be deleted.
4. Financings should be pre-marketed, especially for larger companies, as opposed to "bought deals" which are tantamount to ambushing investors. Pre-marketed deals have the advantage of widening the circle of potential investors. In a favorable market, shares can rise into the offering. We would expect, because of the expenses involved in a premarketed deal, that very small cap and especially pure exploration companies without cash flow would continue to finance in the traditional manner.
5. In any event, all financing circulars should contain detailed use of proceeds schedules.
6. Net present value calculations should not be the primary benchmark for determining whether a proposed acquisition is accretive or dilutive. While it is a useful measure of value, it is based on numerous assumptions about future events, lacks transparency, and can be easily manipulated by bankers and analysts to argue their case.
7. Corporate goals should include a commitment to anti-dilutive financing practices.

The notion that the gold ETF is cannibalizing demand for gold shares is unsubstantiated and circumstantial. It is a convenient myth to explain the apparent lack of responsiveness of gold shares to advances in the gold price. We should note that we are enthusiastic supporters of both mining shares and the ETF, simply because we are bullish on gold. Gold shares remain the preferred vehicle to obtain a leveraged play on the gold price. However, they represent a form of exposure to gold that entails risk not present in ownership of the metal. They are for risk takers and opportunists. Gold itself is for the risk avoiding and safety seeking constituents of the investment universe. Their numbers are far greater than those who would find gold shares appropriate. The ETF is an enabling device to provide risk avoiders easy access to gold. Such a device has never been available in the history of capital markets. Because the population of risk avoiders is vast in comparison to risk takers, we see no reason why the market cap of the ETF in time should not equal or exceed that of gold mining shares. Today, the market cap of the gold ETF is approximately $15 billion versus perhaps $150 billion for gold mining shares. In our opinion, the accessibility to gold provided by the ETF is capable of driving the gold price even in the absence of worst case outcomes imagined by investors when they think of gold. For this potential rising tide, the captains of leaky boats should find much praise.

The essential investment attribute of any gold share is its linkage to a rising gold price. Of course production costs must be effectively managed, especially cash costs. Certainly reserves must be replaced and discoveries must be made. Environmental and social standards must be complied with. Earnings guidance must be provided and quarterly conference calls must be made. All of this and more represent essential day to day activity in the gold mining business. However, the sum of this activity has little or no investment significance unless the company's shares provide an effective connection to the gold price. To assure investors that a strong linkage exists requires that industry leaders provide accountability in terms of appropriate ratios of shares to the physical asset.